Usage

Configuration file (.ini)

The parameters to the solver is given in a configuration file using the INI format.

Each section and available parameters are described below.

[case]

  • v_inf — The inflow velocity in m/s.

  • rpm — Rotational speed of the rotor in rotations per minute.

  • twist — Optional global pitch change in degrees.

  • coaxial — Whether to add a second rotor. Set to False by default.

  • rpm2 — Rotational speed of the second rotor in rotations per minute.

  • twist2 — Optional global pitch change of the second rotor in degrees.

  • dz — Distance to second rotor.

[rotor]

  • nblades — Number of blades.

  • diameter — Diameter in m.

  • radius_hub — Radius of hub in m.

  • section — List of airfoil sections given by airfoil name.

  • radius — List of distance to each section in m.

  • chord — List of chord length of each section in m.

  • pitch — List of pitch of each section in degrees.

[rotor2]

Description of second rotor for coaxial simulations, same parameters as first rotor.

[fluid]

  • rho — Fluid density in kg/m^3.

  • mu — Fluid dynamic viscosity in Pa s.

[solver]

Optional solver settings.

  • solver — The default solver is a bisection solver. This can be replaced by a more stable brute force solver using ‘brute’ here.

  • Cs — Slipstream coefficient for coaxial solver.

Running a single simulation

To run a single simulation, we create a solver object with the configuration file as parameter and then execute the solver.Solver.run() method:

from pybemt.solver import Solver

s = Solver('rotor.ini')
T, Q, P, section_df = s.run()

The solver returns the thrust [N], the torque [Nm] and the power [W]. Additionally, a pandas DataFrame with the result for each rotor section is provided.

Running a parameter sweep

A typical use case is to run a sweep of a parameter, for instance the rotational speed of the rotor. A utility function is provided for this, called solver.Solver.run_sweep().

This is used in the following way:

from pybemt.solver import Solver

s = Solver('rotor.ini')
df, section_df = s.run_sweep('rpm', 20, 150, 350)

The run_sweep function takes as arguments the parameter to sweep, the number of points and the minimum and maximum values. It returns a data frame with values for each parameter value, as well as a list of dataframes for each section.

Running an optimization

Optimizaton of parameters can easily be done using the scipy.optimize library. Currently, only optimization of pitch is supported directly by the library:

from pybemt.solver import Solver

s = Solver('rotor.ini')

pitches = s.optimize_pitch()

The differential evolution algorithm is used in the current implementation, as it has been found to give the best results. Each section is considered as a separate parameter for the optimization. Using a parameterized function instead can lead to significant speedups, but this is not currently directly supported by the package.

Adding a new airfoil

pyBEMT uses the same file format as the AeroDyn software. However, note that currently only the drag and lift tables are used and only a single Reynolds number is supported.

To add a new airfoil, either add the file to the airfoils directory and re-install the software or add the airfoil file directly to the installed airfoils directory. The installation location can be found by running the following code snippet:

import os
import pybemt
print(os.path.dirname(pybemt.__file__))